publication date: Jul. 11, 2014
html Sledge Response.html
By George Sledge
As always, I both enjoyed and learned something interesting from Dr. Don Berry, one of the great biostatisticians of our era. I find, reading his analysis of ALTTO and NeoALTTO, much that I agree with. I am certainly no statistician, and must bow to his statistical analysis of ALTTO and NeoALTTO. And yet at the same time I think it misses the point. So I will beg to differ.
Let us begin with the use of pCR as a predictor of adjuvant benefit. The Cortazar analysis suggests, as does virtually every analysis of neoadjuvant therapy, that having a pathologic complete response is a very good thing for an individual. The great majority of patients with a pCR go on to have prolonged disease-free survival. There is no question but that, for an individual, pCR is a powerful predictor of benefit. This is true in all breast cancer subgroups, and the 0.39 EFS hazard ratio quoted by Dr. Berry refers to this striking relationship.
But that is not the real question for a clinical trialist. The question is, rather (at least in part), what difference in pCR rate must one see between two arms of a neoadjuvant trial for one to see an improvement in event-free survival rates for the overall trial population. The Cortazar meta-analysis was quite clear in this regard: there was no obvious relationship between a delta in pCR and a delta if EFS hazard ratio. This was true even after excluding low-grade, hormone-receptor positive tumors that would be expected not to benefit … Continue reading 40-28 George Sledge Responds To Berry’s Letter: Neoadjuvant Doesn’t Predict Adjuvant In Breast Cancer and Berry’s Rejoinder
To access this members-only content, please log in.
Institutional subscribers, please log in with your IP
If you're not a subscriber why not join today?
To gain access to the members only content click here
You will be given immediate access to premium content on the site.Click here to join.